by Steve Cuno
As seen in the current issue of Deliver magazine
If you have ever touted direct mail’s ability to deliver empirical results, you know about “The Inevitable Question.” Almost on cue, as you rhapsodize about testing, tracking, analysis and predictive results, the client will train an eyeball on you and ask, “What’s a good response rate for direct mail? One percent? Two? Three?”
Framed in that manner, “The Inevitable Question” happens to be a trick question. The direct mail industry holds up no percentage as its universal standard. Rather, a “good” response is one that achieves your objective. If your objective happens to be a response of 1 percent, then no less than 1 percent is a “good” response. If you require 17 percent, “good” starts there.
With that in mind, it might be a good idea to set an objective or two before creating and sending out your direct mail. Obvious? I agree. But I have seen companies launch costly campaigns without regard for such details, only to flounder about in an attempt to measure success afterward. That’s like throwing a dart nowhere in particular and then trying to evaluate your aim.
Other companies wait until results come in and then set their objectives to match. That’s like throwing a dart nowhere in particular, then painting the target where the dart happens to stick and bragging about hitting a bull’s-eye.
Doing the Math
Here’s a saner approach to determining a good response:
1. Project the cost of your direct mail program. Since I’m no math wiz, let’s keep it simple by assuming that you plan to spend $1,000 on printing, addressing, postage, etc.
2. Calculate the profit you will clear, on average, with every response. Again, for simplicity’s sake, let’s say that each response will bring you $100 in net revenue.
3. Aha! You need only 10 responses to break even (10 responses x $100 = $1,000). So if you mailed to 1,000 people, just 1 percent would be a good response. If you mailed to 2,000 people, just one-half of 1 percent would be good. If you mailed to 2,000 people and each respondent made two purchases, then a mere response of one-quarter of 1 percent would be good. And so on.
Is It Possible?
Upon completing the math, the client may say, “Now that I know the response I need, before going much further I’d like to know if that response is within the realm of possibility.”
There are two ways to tackle that one.
Way No. 1: Do the research. Find out if your product or service has been sold by mail before. If so, a track record may exist for grounding expectations in reality. Books, articles and the Internet may yield such information. While you’re at it, start networking and trading information with others in the field. I have had great luck simply by asking direct mail professionals to share their experience.
A caveat is in order: No two products, creative approaches, or sets of market conditions are alike. Even with historical data on your side, it’s wise to test before betting the farm. Which brings us to …
Way No. 2: Test. Send your direct mail to a small but representative, statistically valid sample from your mailing list. If the sample responds at the desired level, repeat the test to another sample. If results hold, you can be reasonably sure they will also hold when you roll out to the rest of your list.
Once you’ve determined what constitutes a good response, remember to clue in the boss ahead of time. Otherwise, “The Inevitable Question” may rear its ugly head after the fact. If your direct mail happens to pull an ROI of two-to-one, the last thing you need is a boss who expresses dismay because it was “… only X percent.”
—Steve Cuno